Free radical induced lipid peroxidation has been indirectly implicated in several types of liver injury including halothane hepatotoxicity. However, current measures of lipid peroxidation are insufficient to evaluate this process, especially when applied to in vivo systems. The recent discovery of non-cyclooxygenase free radical derived prostanoids (F2-isoprostanes) produced in vivo as a result of lipid peroxidation has opened new avenues of investigation. These biologically active prostanoids were detected initially in plasma and levels were shown to increase up to 200-fold in two animal models of severe free radical induced liver injury (CC14 administered to rats and Diquat administered to Se deficient rats). F2- isoprostanes are initially formed from arachidonic acid acylated to phospholipids and quantitation of these compounds in tissues has been accomplished. Liberation of free F2-isoprostanes results from the action of phospholipases on the isoprostane-containing phospholipids. In addition to providing a new measure of lipid peroxidation, these novel compounds suggest at least two possible mechanisms of peroxidation induced damage, disruption of lipid membranes by the modified phospholipids and the biological activity of the liberated prostanoids. Halothane is a commonly used anesthetic which has been associated with both mild and severe hepatic injury. Reductive metabolism of halothane results in production of triflouro-chloroethane radical that can initiate lipid peroxidation. Administration of halothane to phenobarbital induced rats under hypoxic conditions favors deductive metabolism and has been used to study halothane induced liver injury. In this model, lipid peroxidation has usually been assumed without adequate measurement to accompany reductive metabolism. Preliminary results of studies measuring both plasma and liver levels of F~-isoprostanes in this model show substantial increases with halothane administration. Measurements such as these under various conditions of halothane administration resulting, in different degrees of liver damage would allow determination of (1) the relationship between halothane metabolism and non-cyclooxygenase prostanoid formation as a measure of lipid peroxidation, (2) the relationship between lipid peroxidation and hepatic injury in the halothane model. In addition, because reductive metabolism occurs in humans, production of these compounds will be investigated in patients undergoing halothane anesthesia and correlations with hepatic enzyme abnormalities made. The results of this work may provide a model by which similar toxic liver injuries, can be understood and rational therapy designed. In this revised proposal, studies to determine cellular mechanisms in this injury, and the molecular response to it are proposed.